On Wed, 28 Mar 2007 15:35:21 +0200, <james.jones@firstinvestors.com> wrote:

From: "John Maddock" <john@johnmaddock.co.uk>

...

james.jones@firstinvestors.com wrote:

...

From: Jeff Garland <jeff@crystalclearsoftware.com> I have done a lot of work on something very similar to this that I was hoping to propose fairly soon (next few weeks). The interface isn't quite N2198, but I'm not crazy about N2198 anyway (it isn't threadsafe, for example, AFAICT). I've been using the IEEE754r spec. Is it worthwhile at all to continue to work on a BOOST library for fixed decimal arithmetic without making it N2198 compliant?

There would probably be a way to make an N2198-compliant version of my code, if that were necessary.

Good question, I'm very surprised that N2198 isn't thread safe, how come?

I'm not sure about this, actually - but do the fe* functions guarantee thread safety? I read over Sun's documentation and it says they do, but can this generally be assumed? N2198 specifies that cfenv be used for context, and this is where I got concerned.

IMHO thread-safety is more an implementation issue, what let you guess that it is not possible building a thread-safe implementation of interface N2198 ? Marco -- Using Opera's revolutionary e-mail client: http://www.opera.com/mail/